One of the questions I often come back to is this: how much of our collective wealth is inherited by our generations versus created by our generations?

I realized that the keys on the keyboard in front of me might make a good dataset to attack that problem. So I built a small little experiment to explore the history of the keys on my keyboard.

The Five Waves of Symbols

Painting with broad strokes, there were approximately five big waves of inventions that have left their mark on the keyboard. The first wave was the invention of the phonetic alphabet letters. The second wave was the Hindu-Arabic Numerals. The third wave was the mathematical punctuation of the Enlightenment period. The fourth wave was the invention of the typewriter. And the fifth and most recent wave was the invention of the personal computer.

I haven’t made any traditional charts yet with this dataset, but you can roughly make out these waves in the interactive visualization by moving the slider around.

Concentric Circles

An interesting pattern that I never saw before is how the five waves above are roughly arranged in circles. The oldest symbols (letters) are close to the center, followed by the Hindu-Arabic Numbers, surrounded by the punctuation of the Englightenment, surrounded by the keys of the keyboard, surrounded by the recent additions in the P.C. era. Again, painting with broad strokes, but I found that to be an interesting pattern.

Standing on the Shoulders of Giants

All of these waves happened invented before my generation. Almost all of them before any generation alive today. The keyboard dataset provides strong evidence that most of our collective wealth is inherited.

Build Notes

I got this idea last week and couldn’t get it out of my head. Yesterday I took a quick crack at it. I didn’t have much time to spare, just enough to explore the big ideas. I started by typing all the characters on my keyboard into a Tree Notation document. Then I dug up some years for a handful of the symbols. Then I found the great Apple CSS keyboard. I stitched together the two and it seemed to be at least mildly interesting so I opted to continue. I then flushed out most of the dataset. Finally I played around with a number of visualization effects. At first I thought heatmaps would work well, and tried a few variations on that, but wasn’t happy with anything. I posted my work-in-progress to a few friends last night and called it a day. Today I switched to the “disappearing keys” visualization. That definitely felt like a better approach than the heatmap. I made the thing as fun as I could given time constraints and then shipped.

Published 2/25/2020